Improving WalkSAT By Effective Tie-Breaking and Efficient Implementation
Author(s)
Cai, Shaowei
Luo, Chuan
Su, Kaile
Griffith University Author(s)
Year published
2015
Metadata
Show full item recordAbstract
Stochastic local search (SLS) algorithms are well known for their ability to efficiently find models of random instances of the Boolean satisfiability (SAT) problem. One of the most famous SLS algorithms for SAT is WalkSAT, which is an initial algorithm that has wide influence and performs very well on random 3-SAT instances. However, the performance of WalkSAT on random k-SAT instances with k > 3 lags far behind. Indeed, there are limited works on improving SLS algorithms for such instances. This work takes a good step toward this direction. We propose a novel concept namely multilevel make. Based on this concept, we design ...
View more >Stochastic local search (SLS) algorithms are well known for their ability to efficiently find models of random instances of the Boolean satisfiability (SAT) problem. One of the most famous SLS algorithms for SAT is WalkSAT, which is an initial algorithm that has wide influence and performs very well on random 3-SAT instances. However, the performance of WalkSAT on random k-SAT instances with k > 3 lags far behind. Indeed, there are limited works on improving SLS algorithms for such instances. This work takes a good step toward this direction. We propose a novel concept namely multilevel make. Based on this concept, we design a scoring function called linear make, which is utilized to break ties in WalkSAT, leading to a new algorithm called WalkSATlm. Our experimental results show that WalkSATlm improves WalkSAT by orders of magnitude on random k-SAT instances with k > 3 near the phase transition. Additionally, we propose an efficient implementation for WalkSATlm, which leads to a speedup of 100%. We also give some insights on different forms of linear make functions, and show the limitation of the linear make function on random 3-SAT through theoretical analysis.
View less >
View more >Stochastic local search (SLS) algorithms are well known for their ability to efficiently find models of random instances of the Boolean satisfiability (SAT) problem. One of the most famous SLS algorithms for SAT is WalkSAT, which is an initial algorithm that has wide influence and performs very well on random 3-SAT instances. However, the performance of WalkSAT on random k-SAT instances with k > 3 lags far behind. Indeed, there are limited works on improving SLS algorithms for such instances. This work takes a good step toward this direction. We propose a novel concept namely multilevel make. Based on this concept, we design a scoring function called linear make, which is utilized to break ties in WalkSAT, leading to a new algorithm called WalkSATlm. Our experimental results show that WalkSATlm improves WalkSAT by orders of magnitude on random k-SAT instances with k > 3 near the phase transition. Additionally, we propose an efficient implementation for WalkSATlm, which leads to a speedup of 100%. We also give some insights on different forms of linear make functions, and show the limitation of the linear make function on random 3-SAT through theoretical analysis.
View less >
Journal Title
The Computer Journal
Volume
58
Issue
11
Subject
Applied mathematics not elsewhere classified
Information and computing sciences